In the context of airborne Anti-Submarine Warfare (ASW), we focus here on active sonar systems, i.e. systems based on the transmission and reception of an acoustic wave reflected by the target (echo). Furthermore, our case study involves acoustic buoys, or sonobuoys, which can be classified into three main categories: transmitter-only (Tx), receiver-only (Rx) and transmitter-receiver (TxRx). When transmitter and receiver are co-located - TxRx buoy - we refer to this as monostatism, while when transmitter and receiver are located in two distinct geographical locations, we refer to this as bistatism. A Multistatic Sonar Network (MSN) is therefore defined as a combination of sonar systems in monostatic and/or bistatic configuration. In this context, given a limited number of heterogeneous sonobuoys and probabilistic detection models, we seek to determine the optimal MSN, i.e. the one that maximizes the area covered. To address this issue, we propose a constructive greedy heuristic relying on a QuadTree-based approach (QT) enhanced by a local search through Variable Neighborhoud Descent (VND). This heuristic is based on a recursive principle - akin to branch & bound - dividing the area of interest into sectors and accelerating the search for locally optimal positions through judicious computation of upper and lower bounds. Finally, we compare our method with the most efficient Mixed-Integer Linear Program (MILP) in the literature as well as a Simulated Annealing (SA).