Master–Slave Match Line Design for Low-Power Content-Addressable Memory

The analogy of matchline sensing techniques for content addressable memory (CAM)

“…Segmented ML technique proved to be the best to handle all the performance metrics. Table 2 Performance comparison summary of ML-sensing techniques Precharge high [6,14,15] Current-race scheme [22,33] Precharge free [29][30][31][32] Low swing [11,[34][35][36] Segmented [37][38][39][40][41][42][43][44][45] Sele. precharge [46][47][48]…”

“…(iv) Low-swing scheme [11,[34][35][36]: low-swing ML approaches to reduce dynamic power. (v) Segmented ML evaluation [37][38][39][40][41][42][43][44][45]: a non-uniform distribution of ML evaluation results by segmenting ML and designing sections with different lengths and structures to improve the power-delay trade-off. (vi) Selective pre-charge sensing scheme [46][47][48]: another nonuniform power distribution among the MLs based on the higher mismatch probability in initial sections.…”

“…This is because of the following factors: (i) all the mismatching MLs discharge from pre-charged (V DD ) to GND as the MLs are evaluated during the search (all the mismatching MLs change their states from pre-charged as the MLs are evaluated during search). (ii) For a mismatching ML, the number of discharge paths is in linear proportion to the number of mismatching cells [37][38][39][40][41][42][43][44][45]. At most all but one MLs mismatch for a search and hence the MLs are required to be re-pre-charged prior to next search.…”

“…Segmenting the MLs into multiple sections and performing/blocking of evaluation in the partitions separately with different ML structures can save ML power significantly. All these techniques are categorised into six unique types as follows: (i) Pre‐charge high ML sensing technique [6, 14, 15]: a simple yet power hungry approach used by most CAM designers. (ii) Current race sensing [22, 33]: a pre‐charge low strategy on the ML using a current source. (iii) Pre‐charge free ML sensing [29–32]: elimination of extra pre‐charge phase to accommodate more number of searches. (iv) Low‐swing scheme [11, 34–36]: low‐swing ML approaches to reduce dynamic power. (v) Segmented ML evaluation [37–45]: a non‐uniform distribution of ML evaluation results by segmenting ML and designing sections with different lengths and structures to improve the power‐delay trade‐off. (vi) Selective pre‐charge sensing scheme [46–48]: another non‐uniform power distribution among the MLs based on the higher mismatch probability in initial sections. …”

Low power CMOS based Self Controlled Precharge Free Content Addressable Memory with Minimum area for Image Processing Devices

“…Another drawback of the conventional Cam is that the frequent switching of ML causes the switching power to increase. This switching can avoided by the segmentation of words into several small words or the Master -Slave architecture can be implemented so that the first segment or master is matched then only the remaining part is involved in the search operation [4]. The next invention to reduce the drawback of the above method is that the Pre-Computation in which each stored word is having parity Bit and search starts with the matching of parity bit of both stored and search word if it matches then the original word can searched otherwise the process can be stopped there itself [5].…”