【多视角】Multi-Point Integrated Sensing and Communication: Fusion Model and Functionality Selection

最新推荐文章于 2024-10-02 19:15:37 发布

db_1024

最新推荐文章于 2024-10-02 19:15:37 发布

阅读量740

点赞数 11

文章标签：人工智能算法机器学习

本文链接：https://blog.csdn.net/qq_41100635/article/details/142639783

版权

                    
                        
                    
                    Use  K K K dual-functional radars (DFR) to sense one target object and communicate with one user.
A functionality selection module, which is represented by a binary variable  x = \mathbf{x}= x=  [ x 1 , … , x K ] T ∈ { 0 , 1 } K [x_1,\ldots,x_K]^T\in\{0,1\}^K [x1​,…,xK​]T∈{
       0,1}K,is adopted to determine the working states of all the DFRs. Specifically,  x i = 1 x_i=1 xi​=1,  ∀ i ∈ { 1 , … , K } \forall i\in\{1,\ldots,K\} ∀i∈{
       1,…,K}, represents that the  i i i-DFR operates in the sensing mode and  x i = 0 x_i=0 xi​=0 denotes that the  i i i-DFR operates in the communication mode.
we aim to to use the target information of sensing signal for classification problem. Effective DFRs are all the sensing DFRs, i.e.  x i = 1 x_i=1 xi​=1. Meanwhile, SINR of sensing signal of effective DFRs should be greater than a threshold  γ \gamma γ. The set of effective DFRs is thus  E = { i : x i = 1 \mathcal{E}=\{i:x_i=1 E={
       i:xi​=1, SINR i s ≥ γ } _i^s\geq\gamma\} is​≥γ} whose cardinality is  ∣ E ∣ . |\mathcal{E}|. ∣E∣.
we aim to fuse the outputs of the effective DFRs  E \mathcal{E} E via voting.
Fusion accuracy with the voting threshold  n n n is  Θ ( E ∣ n ) = 1 2 ∑ l = 0 n − 1 ∑ F ∈ F l ∏ i ∈ F P i ∏ i ∈ F c ( 1 − P i ) + 1 2 ∑ l = 0 ∣ E ∣ − n ∑ F ∈ F l ∏ i ∈ F Q i ∏ i ∈ F c ( 1 − Q i ) \Theta(\mathcal{E}|n)=\frac12\sum_{l=0}^{n-1}\sum_{F\in\mathcal{F}_l}\prod_{i\in F}P_i\prod_{i\in F^c}(1-P_i)+\frac12\sum_{l=0}^{|\mathcal{E}|-n}\sum_{F\in\mathcal{F}_l}\prod_{i\in F}Q_i\prod_{i\in F^c}(1-Q_i) Θ(E∣n)=21​∑l=0n−1​∑F∈Fl​​∏i∈F​Pi​∏i∈Fc​(1−Pi​)+21​∑l=0∣E∣−n​∑F∈Fl​​∏i∈F​Qi​∏i∈Fc​(1−Qi​), where  P i P_i Pi​ and  Q i Q_i Qi​ are the false negative (false alarm) and false positive (missing alarm) rates at the  i i i-th DFR, respectively, are estimated from the experimental data; note that different DFRs may have different  ( P i , Q i ) (P_i,Q_i) (Pi​,Qi​) under the same SINR due to different observation angles; (2)  F l \mathcal{F}_l Fl​ contains all the  l l s u b s e t s   w i t h   l u n i q u e   D F R s   f r o m   E a n d   i t s   c a r d i n a l i t y   i s ∣ F l ∣ = ( ∣ E ∣ l ) = ∣ E ∣ ! l ! ( ∣ E ∣ − l ) ! ; ( 3 )  For any  F ∈ F l ,  its comple- {ll}\mathrm{subsets~with~}l\mathrm{unique~DFRs~from~}\mathcal{E}\mathrm{and~its~cardinality~is}\\|\mathcal{F}_l|=(\frac{|\mathcal E|}l)=\frac{|\mathcal E|!}{l!(|\mathcal E|-l)!};(3)\text{ For any }F\in\mathcal{F}_l,\text{ its comple-} llsubsets with lunique DFRs from Eand its cardinality is∣Fl​∣=(l∣E∣​)=l!(∣E∣−l)!∣E∣!​;(3) For any F∈Fl​, its comple- ment is denoted as  F c = E \ F . F^c=\mathcal{E}\backslash F. Fc=E\F. It can be seen that the fusion accuracy under optimal voting is  max ⁡ n Θ ( E ∣ n ) . \max_n\Theta(\mathcal{E}|n). max

                