fix in gnb mod_dci buffer size
This commit is contained in:
@@ -162,7 +162,7 @@ static void nr_generate_dci(PHY_VARS_gNB *gNB,
|
||||
scrambled_output[6], scrambled_output[7], scrambled_output[8], scrambled_output[9], scrambled_output[10],scrambled_output[11] );
|
||||
#endif
|
||||
/// QPSK modulation
|
||||
c16_t mod_dci[NR_MAX_DCI_SIZE >> 2] __attribute__((aligned(16)));
|
||||
c16_t mod_dci[NR_MAX_DCI_SIZE / 2] __attribute__((aligned(16)));
|
||||
nr_modulation(scrambled_output, encoded_length, DMRS_MOD_ORDER, (int16_t *)mod_dci); // Qm = 2 as DMRS is QPSK modulated
|
||||
#ifdef DEBUG_DCI
|
||||
|
||||
|
||||
Reference in New Issue
Block a user