We propose a novel hybrid combination scheme in cooperative spectrum sensing (CSS), which utilizes the diversity of reporting channels to achieve better throughput performance. Secondary users (SUs) with good reporting channel quality transmit quantized local observation statistics to fusion center (FC), while others report their local decisions. FC makes the final decision by carrying out hybrid combination. We derive the closed-form expressions of throughput and detection performance as a function of the number of SUs which report local observation statistics. The simulation and numerical results show that the hybrid combination scheme can achieve better throughput performance than hard combination scheme and soft combination scheme.
Cognitive radio (CR) is a promising technology to combat spectrum scarcity by exploiting the unused spectrum bands licensed to the primary user (PU) [
There are some researches which investigate overhead-throughput tradeoff in cognitive radio network (CRN) that consists of a number of the SUs employing energy detectors and a single decision FC. Many methods have been proposed to reduce the reporting overhead. Some introduce multibit quantization combining scheme [
In this letter, we propose a novel hybrid combination scheme to achieve better throughput performance. SU transmits either local observation statistics or local decision to FC. FC makes the final decision by carrying out hybrid combination. The theoretical analysis and simulation show that the hybrid combination scheme outperforms hard combination and soft combination scheme.
We consider the scenario illustrated in Figure
System model of hybrid combination scheme.
The frame structure for the periodic spectrum sensing is shown in Figure
Frame structure of hybrid combination scheme.
Local spectrum sensing problem using energy detection can be formulated as the following binary hypothesis test:
The probability of local detection is given by
In the hybrid combination system model,
Then the probability of local detection
By substituting (
For
For large
Then we can get
According to (
By (
We assume that the PU is either absent or present during each frame duration (Figure
We assume that the local observation statistics are quantized with
The throughput is given by
Intuitively, there is a tradeoff between
The probability of detection versus probability of false alarm with different
The system is set up as follows: the number of SUs in the CRN is set to be
Figure
In addition, the performance grows better when
Figure
The false alarm probability versus the number of SUs
The results in Figures
The throughput versus the number of SUs
Performance comparison between the proposed hybrid combination scheme and the exiting scheme with different reporting rate vectors.
Figure
No. | Reporting rate vector (RRV) (kbit/s) |
---|---|
1 |
|
2 |
|
3 |
|
Figure
No. | Reporting rate vector (RRV) (kbit/s) |
---|---|
1 |
|
2 |
|
3 |
|
4 |
|
5 |
|
6 |
|
7 |
|
With different RRVs, the result of comparison between soft and hard combination scheme varies, which means the diversity of reporting channel quality has great influence on the choice of combination scheme.
However, the result shows that the proposed hybrid combination scheme outperforms both hard combination scheme [
According to
In this paper, we investigated overhead-throughput tradeoff over different reporting channel quality in CRN that consists of a number of the SUs employing energy detectors and a single decision fusion center. We proposed a novel hybrid combination scheme with low overhead and complexity, which outperforms both hard combination and soft combination scheme, especially when the diversity of the reporting channel is obvious.
The authors declare that they have no conflict of interests regarding the publication of this paper.
This work was supported by the National Natural Science Foundation of China (Grant nos. 61172062, 61301160), Jiangsu Province Natural Science Foundation (Grant no. BK2011116), and in part by the National Basic Research Program of China (Grant no. 2009CB320400).