Detection of wireless signals in broadband cognitive PLC using the modified 2-D LAD ACC algorithm

There is a need to fulfill the growing capacity requirements of power line communication (PLC). One way to achieve higher data rates in PLC networks is to operate at higher frequencies and increase the used bandwidth. In the future, the increased PLC frequency range can overlap with several wireless communications systems resulting in interference for these transmissions and vice versa. Commonly used static notching results in inefficient spectrum utilization as in a given place and at a given time only a part of the frequency range may be occupied by other wireless systems. Some wireless signals like FM signals are very closely spaced. In practice it is impossible to utilize these narrowband unoccupied frequency bands. In this case we can cluster closely spaced signals. Guard bands around detected wireless signals must also be taken into account before defining usable frequency bands for PLC. In this paper, we use the 2-D LAD ACC algorithm to detect and cluster interfering signals. We also further develop the algorithm so that after clustering guard bands are added and reclustering, if necessary, is executed. The performance of this modified 2-D LAD ACC algorithm is analyzed using calculations and spectrum analyzer measurement results obtained from in-office PLC network in the frequency range 80-200 MHz. The results verify that the proposed method performs well by detecting and clustering continuous and non-continuous wireless signals and providing means for clever spectrum utilization in PLC-network.