METHOD AND SYSTEM FOR OPTIMIZING TRANSCEIVER SPECTRUM SHARING

Abstract

A method and system for providing a cooperative spectrum sharing model that jointly optimizes primary user equipment parameters for improved frequency agility and performance while mitigating mutual interference between the primary user equipment and secondary user equipment. Spectrum sensing is implemented to form a power spectral estimate of the electromagnetic environment (EME) and apply multi-objective optimization to adjust the operational parameters of the primary user equipment to mitigate interference.

Claims

1. A system for optimizing radio frequency (RF) spectrum sharing comprising: a radar system; a RF communication system configured to transmit and receive signals to user equipment; a spectrum sensing system, coupled to the radar system and the RF communication system, and comprising at least one computer processor configured to sense the RF spectral environment proximate the radar system and the RF communication system, process the sensed RF spectral environment using multi-obj ective optimization to jointly determine an optimal operational RF spectrum for each of the radar system and the RF communication system and control operational parameters of the radar system and the RF communication system to optimize RF spectrum sharing amongst the radar system and the RF communication system.

2. The system of claim 1 wherein the operational parameters include at least one of power output, frequency, or bandwidth.

3. The system of claim 1 wherein the processing utilizes multi-obj ective processing to maximize a first objective function of radar signal to interference plus noise ratio (SINR), a second objective function of radar range resolution, and a third objective function of radio capacity of the user equipment.

4. The system of claim 3 wherein multi-objective processing is performed using a genetic algorithm.

5. The system of claim 4 wherein the genetic algorithm is performed using a Multi-objective Genetic Algorithm (MOGA), Niched Pareto Genetic Algorithm (NPGA), Weight-based Genetic Algorithm (WBGA), Random Weighted Genetic Algorithm (RWGA), Nondominated Sorting Genetic Algorithm (NSGA), Strength Pareto Evolutionary Algorithm (SPEA), improved SPEA (SPEA2), Pareto-Archived Evolution Strategy (PAES), Pareto Envelope-based Selection Algorithm (PESA), Region-based Selection in Evolutionary Multiobjective Optimization (PESA-II), Fast Nondominated Sorting Genetic Algorithm (NSGA-II), Multi-objective Evolutionary Algorithm (MEA), Micro-GA, Rank-Density Based Genetic Algorithm (RDGA), or Dynamic Multi-objective Evolutionary Algorithm (DMOEA) approach.

6. (canceled)

7. A method of optimizing radio frequency (RF) spectrum sharing between a radar system and a RF communication system configured to transmit and receive RF signals to user equipment, the method comprising: sensing a RF spectral environment proximate the radar system and the RF communication system; processing the sensed RF spectral environment using multi-objective optimization to jointly determine an optimal operational RF spectrum for each of the radar system and the RF communication system; and controlling operational parameters of the radar system and the RF communication system to optimize RF spectrum sharing amongst the radar system and the RF communication system.

8. The method of claim 7 wherein the operational parameters include at least one of power output, frequency, or bandwidth.

9. The method of claim 7 wherein the processing further comprises multi-objective processing to maximize a first objective function of radar signal to interference plus noise ratio (SINR), a second objective function of radar range resolution, and a third objective function of radio capacity of the user equipment.

10. The system of claim 9 wherein multi-objective processing comprises executing a genetic algorithm.

11. The system of claim 10 wherein the genetic algorithm is a Multi-objective Genetic Algorithm (MOGA), Niched Pareto Genetic Algorithm (NPGA), Weight-based Genetic Algorithm (WBGA), Random Weighted Genetic Algorithm (RWGA), Nondominated Sorting Genetic Algorithm (NSGA), Strength Pareto Evolutionary Algorithm (SPEA), improved SPEA (SPEA2), Pareto-Archived Evolution Strategy (PAES), Pareto Envelope-based Selection Algorithm (PESA), Region-based Selection in Evolutionary Multiobjective Optimization (PESA-II), Fast Nondominated Sorting Genetic Algorithm (NSGA-II), Multi-objective Evolutionary Algorithm (MEA), Micro-GA, Rank-Density Based Genetic Algorithm (RDGA), or Dynamic Multi-objective Evolutionary Algorithm (DMOEA) approach.

12. (canceled)

13. A non-transitory computer readable medium having software instructions that, when executed by at least one computer processor, perform a method of optimizing radio frequency (RF) spectrum sharing between a radar system and a RF communication system which is configured to transmit and receive RF signals to user equipment, the method comprising: sensing a RF spectral environment proximate the radar system and the RF communication system; processing the sensed RF spectral environment using multi-objective optimization to jointly determine an optimal operational RF spectrum for each of the radar system and the RF communication system; and controlling operational parameters of the radar system and the RF communication system to optimize RF spectrum sharing amongst the radar system and the RF communication system.

14. The method of claim 13 wherein the operational parameters include at least one of power output, frequency, or bandwidth.

15. The method of claim 13 wherein the processing further comprises multi-objective processing to maximize a first objective function of radar signal to interference plus noise ratio (SINK), a second objective function of radar range resolution, and a third objective function of radio capacity of the user equipment.

16. The method of claim 15 wherein multi-objective processing comprises executing a genetic algorithm.

17. The method of claim 16 wherein the genetic algorithm is Multi-objective Genetic Algorithm (MOGA), Niched Pareto Genetic Algorithm (NPGA), Weight-based Genetic Algorithm (WBGA), Random Weighted Genetic Algorithm (RWGA), Nondominated Sorting Genetic Algorithm (NSGA), Strength Pareto Evolutionary Algorithm (SPEA), improved SPEA (SPEA2), Pareto-Archived Evolution Strategy (PAES), Pareto Envelope-based Selection Algorithm (PESA), Region-based Selection in Evolutionary Multiobjective Optimization (PESA-II), Fast Nondominated Sorting Genetic Algorithm (NSGA-II), Multi-objective Evolutionary Algorithm (MEA), Micro-GA, Rank-Density Based Genetic Algorithm (RDGA), or Dynamic Multi-objective Evolutionary Algorithm (DMOEA) approach.

18. The method of claim 13 wherein controlling the operational parameters comprises modifying the center frequency, bandwidth and peak transmit power of the primary user equipment.

19. The system of claim 3, wherein the first objective function of the radar SINR is defined according to Equation (1); the second objective function of the radar range resolution is defined according to Equation (4); and third objective function of the user equipment capacity is defined according to Equation (5).

20. The method of claim 9, wherein the first objective function of the radar SINR is defined according to Equation (1); the second objective function of the radar range resolution is defined according to Equation (4); and third objective function of the user equipment capacity is defined according to Equation (5).

21. The method of claim 15, wherein the first objective function of the radar SINR is defined according to Equation (1); the second objective function of the radar range resolution is defined according to Equation (4); and third objective function of the user equipment capacity is defined according to Equation (5).

22. The system of claim 1, wherein the user equipment comprises primary user equipment and secondary user equipment.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0008] So that the manner in which the above recited features of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to embodiments, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered limiting of its scope, for the invention may admit to other equally effective embodiments.

[0009] FIG. 1 depicts a scenario for co-operative spectrum sharing between primary user equipment and secondary user equipment;

[0010] FIG. 2 depicts an exemplary power spectrum of spectral activity in an exemplary scenario as depicted in FIG. 1;

[0011] FIG. 3 depicts a block diagram of one embodiment of a spectrum sensing system of FIG. 1; and

[0012] FIG. 4 is a flow diagram of an optimization method used by embodiments of the present invention.

DETAILED DESCRIPTION

[0013] Embodiments of the invention include a method and system for optimizing spectrum sharing among transceivers. In one embodiment, the system comprises a spectrum sensing system that is coupled to certain controllable transceivers (primary user equipment) such as communications systems and radar systems. The spectrum sensing system monitors a relevant spectrum for background interference, secondary user equipment transmissions, and primary user equipment transmissions. An optimization method analyses the spectrum and adapts the utilization of the spectrum by the primary user equipment to optimize sharing of the spectrum with the secondary users.

[0014] The spectrum sharing scenario 100, illustrated in FIG. 1, depicts a tracking radar 102 operating within the same vicinity of a communications base station (CBS) 106 and user equipment (UE) 104. The radar 102 is ground-based and tracks a generic airborne moving target 108. Embodiments of the invention is modifiable to support different communication system types (LTE, Global System for Mobile (GSM) communication, 5G, etc). Together the radar 102 and communication base station 106 represent primary user equipment 114 with equal rights to the spectrum and can coordinate their spectrum access to maximize performance. The UE 194 represent secondary user equipment 116 that must be permitted to share the spectrum of the primary user equipment 114.

[0015] The radar 102 and target 108 are located at positions Pi and Po, respectively, separated by a distance of R.sub.10. The CBS 106 and UE 104 are located at positions P.sub.2 and P.sub.3, respectively, separated by a distance of R.sub.23. R.sub.13 indicates the distance between the radar 102 and the UE 104, while R.sub.21 indicates the distance between the radar 102 and the CBS 106. In this scenario, the capacity of the downlink channel is examined, and the UE 104 is positioned at the minimal separation distance to the radar 102 (close as possible), denoted as R13, within the main beam of the radar 102. This distance represents the maximum interference possible from the radar 102 to the UE 104, i.e., the worst-case scenario.

[0016] The scenario 100 of FIG. 1 also illustrates RF interference from secondary user equipment 116, where these emitters are considered secondary users whose goal is to access underutilized spectrum. Interference from the primary users to the secondary users is therefore allowed, but interference from the secondary users to the primary users should be mitigated. However, given the need for co-existence, the primary users should limit their spectrum access and mitigate mutual interference, if possible. For example, it is possible for the radar to reduce its bandwidth in order to operate in a channel with high SINR when no target is present. Spectrum sharing is therefore established between the primary and secondary user equipment 114 and 116. Then, after a target 108 is detected, the radar 102 can trade-off SINR for more bandwidth (as the target becomes closer to the radar) to maximize its range resolution while maintaining target detection.

[0017] A spectrum sensing system 110 shown in FIG. 1 generates knowledge of the secondary user activity. Specifically, the spectrum sensing system 110 uses one or more antennas 112 to passively sense the electromagnetic environment (EME). The system 110 estimates a power spectrum of the EME that is defined as ={.sub.1, . . . , .sub.N} of size N for the baseband frequencies F={f.sub.1, . . . , f.sub.N} with bandwidth B and frequency resolution .sub.B, where .sub.n is the n.sup.th frequency domain sample. For example, consider the spectrum in FIG. 2 that is used here for illustration purposes. This spectrum was recorded with B=100 MHz and .sub.B=100 kHz. The spectrum sensing system analyzes the spectrum by implementing multi-objective optimization (MO) to determine the best channel, or sub-band, for radar and communication system dynamic spectrum access (DSA). Note that it is possible for radar or the communication system to occupy more bandwidth and radiate within the same band of the secondary users. It is also possible for radar and the communication system to occupy the same sub-band with a reduced transmit power. The consequence for either of these solutions results in mutual interference that reduces the SINR of all systems. Analysis of the spectrum is therefore necessary to determine the optimal trade-off solution. The optimal sub-band information is then transferred to the radar and the CBS such that the transceiver parameters are adjusted to optimize interference mitigation. The optimal sub-band information is sent to the radar and CBS via wired or wireless communications. In this manner, the spectrum sensing system, in effect, controls the operational parameters of the primary user equipment.

[0018] The SS-MO solution is found using a multiobjective genetic algorithm as, for example, described in A. Konak, D. Coitb, and A. Smith, Multi-Objective Optimization Using Genetic Algorithms: A Tutorial, Reliability, Engineering, and System Safety, vol. 91, no. 9, pp. 992-1007, September 2006, hereby incorporated by reference in its entirety. There are many genetic algorithms that may find use in various embodiments of the invention including, but not limited to: Multi-objective Genetic Algorithm (MOGA), Niched Pareto Genetic Algorithm (NPGA), Weight-based Genetic Algorithm (WBGA), Random Weighted Genetic Algorithm (RWGA), Nondominated Sorting Genetic Algorithm (NSGA), Strength Pareto Evolutionary Algorithm (SPEA), improved SPEA (SPEA2), Pareto-Archived Evolution Strategy (PAES), Pareto Envelope-based Selection Algorithm (PESA), Region-based Selection in Evolutionary Multiobjective Optimization (PESA-II), Fast Nondominated Sorting Genetic Algorithm (NSGA-II), Multi-objective Evolutionary Algorithm (MEA), Micro-GA, Rank-Density Based Genetic Algorithm (RDGA), and Dynamic Multi-objective Evolutionary Algorithm (DMOEA).

[0019] One specific example of a genetic algorithm is the NSGA-II technique described in K. Deb, A. Pratap, S. Agarwal, T. Meyarivan, A fast and elitist multiobjective genetic algorithm: NSGA-II, IEEE Transactions on Evolutionary Computation, vol. 6, no. 2, pp. 182-197, April 2002, hereby incorporated by reference in its entirety. NSGA-II sorts a population of individuals, where each individual represents a decision variable with a corresponding solution. The solution for each individual is found using objective functions. After an initial parent population is randomly generated, NSGA-II implements: 1) crossover and mutation; 2) a formation of an elite population; and 3) a population sort and rank procedure. Each iteration of this procedure produces the next generation of samples, i.e., the children, within the population. Simulated binary crossover (SBX) is used for the crossover procedure with parameter .sub.c. A large .sub.c produces children very similar to the parents, where a small .sub.c produces dissimilar children. Polynomial mutation is used for the mutation process with parameter .sub.m, a variable that controls the similarity between the original and mutated individual (with properties similar to .sub.c). An elite population of M individuals is then formed by combining the parent and child generations, which is then sorted and ranked using the non-dominated procedure. The goal of the genetic algorithm is to evolve this population over T generations such that the decision variables converge to the optimal solution.

[0020] The decision variable vector is defined as x={x.sub.1, x.sub.2, x.sub.3, x.sub.4, x.sub.5, x.sub.6}. The variable x.sub.1=P.sub.1 is the radar transmitter power, where 0P.sub.1P.sub.1,max and P.sub.1,max is the max available power. The radar bandwidth is defined as x.sub.2=.sub.1(i)=.sub.B, where i {1, . . . N}. Note that .sub.1(i)=B, the full bandwidth solution, when i=N and .sub.1(i)=B, the frequency resolution, when i=1. x.sub.3=.sub.1(j)=f.sub.j F is the lower, or start, frequency of the linear frequency modulated (LFM) waveform, where j {1, . . . N}. The lower frequency is used in this development, as opposed to the center frequency .sub.1(j)+.sub.1(j) l 2, to make the mathematical development more convenient. Note that .sub.1(i)+.sub.1(j)B, i.e., the operational band of the radar cannot exceed the upper limit of the baseband. The CBS transmitter power is defined as x.sub.4=P.sub.2, where 0P.sub.2P.sub.2,max and P.sub.2,max is the max available power. x.sub.5=.sub.2(k) is the CBS and UE bandwidth of operation, where k {1, . . . K}. The variable x.sub.6=.sub.2(l)=f.sub.l F is the lower frequency of the CBS and UE bandwidth of operation, where l {1, . . . N}.

[0021] The radar SINR objective function is defined as

Z.sub.1=P.sub.1G.sub.1.sup.2.sup.2N.sub.P.sub.1(i)/[L.sub.1(4).sup.3R.sub.10.sup.4(l.sub.21(P.sub.2, i,j,k,l)+.sub.1(i,j))](1)

[0022] where G.sub.1 is the radar antenna gain, is wavelength, is the target radar cross section, N.sub.P is the number of pulses per coherent processing interval (CPI), L.sub.1 is the radar system loss, is pulse width, and .sub.1>100 is the time-bandwidth product for the linear frequency modulated waveform. The variable

[00001]$\begin{array}{cc}{}_{1\left(i,j\right)}=\{\begin{array}{cc}{}_j,& i=1,j=1,\mathrm{.Math.},N\\ {}_1\left(1,j\right)+{}_1\left(1,j+1\right)& i=2,j=1,\mathrm{.Math.},N-1\\ {}_1\left(i-1,j\right)+{}_1\left(1,i+j-1\right)& i=3,\mathrm{.Math.}N,j=1,\mathrm{.Math.},N-i+1\end{array}& \left(2\right)\end{array}$

is the interference and noise power estimate for all contiguous sub-bands in the spectrum produced by the secondary users as seen by the radar. The radar receiver noise factor is defined as N.sub.f1, but the receiver noise power is inherently estimated in (2) by summing the noise floor for different bandwidth combinations. The interference from the eNodeB to the radar is defined as

l.sub.21(P.sub.2,i,j,j,l)=P.sub.2 G.sub.1 G.sub.2 .sub.21/FDR(i, j, k, l) (3)

[0023] where G.sub.2 is the CBS antenna gain, .sub.21 is the path loss between the CBS and the radar, and FDR(i, j, k, l) is the Frequency Dependent Rejection (FDR) that measures the interference rejection between the radar and CBS. The FDR offset is based on the co-channel and adjacent channel interference between the two systems. Only co-channel interference is of interest. Note that FDR(i,j,k,l) is dependent on the decision variables, hence more interference occurs when the operating sub-bands of the radar and eNodeB overlap.

[0024] The second objective function is the radar range resolution defined as

.sub.R=c/[2 .sub.1(i)], (4)

where c is the speed of light. A small resolution cell size is advantageous for separating closely spaced point targets in range or extracting features from extended targets. Ideally, the radar would occupy .sub.1(N)=B in order to maximize (4), but that decision would decrease (1) (SINR) due to the interference generated by (2) and (3).

[0025] The final objective function is the UE capacity modeled as

Z.sub.3=.sub.2(k) log.sub.2[1+.sub.3], (5)

where

.sub.3=P.sub.2 G.sub.2 G.sub.3 /[L.sub.2(l.sub.13(P.sub.1,i,j)+(k,l))](6)

is the SINR of the UE 104. G.sub.3 is the antenna gain of the UE 104 and .sub.23 is the path loss between the CBS 106 and the UE 104. The variable

[00002]$\begin{array}{cc}{}_{2\left(k,l\right)}=\{\begin{array}{cc}{}_j,& i=1,j=1,\mathrm{.Math.},N\\ {}_2\left(1,l\right)+{}_1\left(1,l+1\right)& i=2,j=1,\mathrm{.Math.},N-1\\ {}_2\left(k-1,l\right)+{}_2\left(1,k+l-1\right)& i=3,\mathrm{.Math.}N,j=1,\mathrm{.Math.},N-i+1\end{array}& \left(7\right)\end{array}$

is the interference and noise power estimate for all contiguous sub-bands in the spectrum produced by the secondary RF emitters as seen by the UE 104. The UE noise factor is defined as N.sub.f3. The interference from the radar to the UE 104 is defined as

l.sub.13(P.sub.1,i,j)=P.sub.1 G.sub.1 G.sub.3 .sub.13/[L.sub.1 L.sub.3 FDR(i, j, k, l)](8)

where L.sub.3 is the UE system loss and .sub.13 is the path loss between the radar and the UE.

[0026] The goal of the NSGA-II approach is to find the decision vector x*={x.sub.1*, x.sub.2*, x.sub.3*, x.sub.4*, x.sub.5*, x.sub.6*} that maximizes the objective functions in (1), (4), and (6):

Z(x*)={Z.sub.1(x*), Z.sub.2(x*), Z.sub.3(x*)}(9)

in the solution space X subject to the constraints Z.sub.1(x*)Z.sub.1,min and Z.sub.2(x*)Z.sub.2,min, and Z.sub.3(x*)Z.sub.3,min, where Z.sub.1,min, Z.sub.2,min, and Z.sub.3,min are the boundary conditions for minimum SINR, bandwidth, and capacity respectively. The solution in (9) is considered feasible if it satisfies these boundary conditions

[0027] FIG. 3 depicts a block diagram of one embodiment of the spectrum sensing system 110. The system 110 comprises at least one central processing unit (CPU) 302 that is capable of performing the signal processing described above. Such a CPU 302 may include one or more microprocessors, gate arrays, microcontrollers, and the like. Functionality of the CPU is supported by support circuits 304 that may include power supplies, clock circuits, input/output circuit(s), analog to digital converter(s), filters and other well-known support circuits. The support circuits may provide sampling, filtering or other pre-processing of the signals received by the antennas 112. A memory 306 (also referred to as a computer readable medium) stores software 308 that is executed by the CPU 302. The memory may comprise a combination of random access memory, read only memory, removable storage, hard drive storage, solid state storage, and the like. The software 308 specifically comprises multi-objective processing software that, when executed by the CPU 302, performs the multi-objective processing on the signals received by the antennas 112 as described above as well as described below with reference to FIG. 4.

[0028] FIG. 4 depicts a flow diagram of a method 400 for optimizing spectrum sharing in accordance with one embodiment of the invention. In one embodiment of the invention, the method 400 is performed in part by executing the software 308 referred to in FIG. 3. The method 400 begins, at step 402, by sensing the surrounding spectrum representing the EME. At step 404, the sensed spectrum is processed as described above using the NSGA-II approach. At step 406, the outcome of the spectrum processing is used to control the primary user equipment such that the utilization of the spectrum is optimized, i.e., primary and secondary users share the spectrum with little impact on performance. Primary user equipment parameters to be controlled include, but are not limited to, transmission power, bandwidth used, frequency allocation for transmission and/or reception, and the like.

[0029] Aspects of this invention have been previously disclosed by the inventors in a paper titled Joint Radar and Communication System Optimization for Spectrum Sharing, which was presented at the 2019 IEEE Radar Conference, Boston Mass., 22-26 Apr. 2019. This paper is herein incorporated by reference in its entirety.

[0030] While the foregoing is directed to embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.